This paper discusses particle filtering in general hidden Markov models (HMMs) and presents novel theoretical results on the long-term stability of bootstrap-type particle filters. More specifically, we establish that the asymptotic variance of the Monte Carlo estimates produced by the bootstrap filter is uniformly bounded in time. On the contrary to most previous results of this type, which in general presuppose that the state space of the hidden state process is compact (an assumption that is rarely satisfied in practice), our very mild assumptions are satisfied for a large class of HMMs with possibly noncompact state space. In addition, we derive a similar time uniform bound on the asymptotic $\mathsf{L}^{p}$ error. Importantly, our results hold for misspecified models; that is, we do not at all assume that the data entering into the particle filter originate from the model governing the dynamics of the particles or not even from an HMM.
Ann. Appl. Probab.
24(5):
1767-1802
(October 2014).
DOI: 10.1214/13-AAP962
[1] Bain, A. and Crisan, D. (2009). Fundamentals of Stochastic Filtering. Stochastic Modelling and Applied Probability 60. Springer, New York. MR2454694[1] Bain, A. and Crisan, D. (2009). Fundamentals of Stochastic Filtering. Stochastic Modelling and Applied Probability 60. Springer, New York. MR2454694
[3] Chopin, N. (2002). A sequential particle filter method for static models. Biometrika 89 539–551. MR1929161 10.1093/biomet/89.3.539[3] Chopin, N. (2002). A sequential particle filter method for static models. Biometrika 89 539–551. MR1929161 10.1093/biomet/89.3.539
[4] Crisan, D. and Heine, K. (2008). Stability of the discrete time filter in terms of the tails of noise distributions. J. Lond. Math. Soc. (2) 78 441–458. MR2439634 10.1112/jlms/jdn032[4] Crisan, D. and Heine, K. (2008). Stability of the discrete time filter in terms of the tails of noise distributions. J. Lond. Math. Soc. (2) 78 441–458. MR2439634 10.1112/jlms/jdn032
[6] Del Moral, P. and Guionnet, A. (1999). Central limit theorem for nonlinear filtering and interacting particle systems. Ann. Appl. Probab. 9 275–297. MR1687359 10.1214/aoap/1029962742 euclid.aoap/1029962742
[6] Del Moral, P. and Guionnet, A. (1999). Central limit theorem for nonlinear filtering and interacting particle systems. Ann. Appl. Probab. 9 275–297. MR1687359 10.1214/aoap/1029962742 euclid.aoap/1029962742
[7] Del Moral, P. and Guionnet, A. (2001). On the stability of interacting processes with applications to filtering and genetic algorithms. Ann. Inst. Henri Poincaré Probab. Stat. 37 155–194. MR1819122 10.1016/S0246-0203(00)01064-5[7] Del Moral, P. and Guionnet, A. (2001). On the stability of interacting processes with applications to filtering and genetic algorithms. Ann. Inst. Henri Poincaré Probab. Stat. 37 155–194. MR1819122 10.1016/S0246-0203(00)01064-5
[8] Del Moral, P., Jacod, J. and Protter, P. (2001). The Monte-Carlo method for filtering with discrete-time observations. Probab. Theory Related Fields 120 346–368. MR1843179 10.1007/PL00008786[8] Del Moral, P., Jacod, J. and Protter, P. (2001). The Monte-Carlo method for filtering with discrete-time observations. Probab. Theory Related Fields 120 346–368. MR1843179 10.1007/PL00008786
[9] Del Moral, P. and Ledoux, M. (2000). Convergence of empirical processes for interacting particle systems with applications to nonlinear filtering. J. Theoret. Probab. 13 225–257. MR1744985 10.1023/A:1007743111861[9] Del Moral, P. and Ledoux, M. (2000). Convergence of empirical processes for interacting particle systems with applications to nonlinear filtering. J. Theoret. Probab. 13 225–257. MR1744985 10.1023/A:1007743111861
[10] Del Moral, P. and Miclo, L. (2000). Branching and interacting particle systems approximations of Feynman–Kac formulae with applications to nonlinear filtering. In Séminaire de Probabilités, XXXIV. Lecture Notes in Math. 1729 1–145. Springer, Berlin. MR1768060 10.1007/BFb0103798[10] Del Moral, P. and Miclo, L. (2000). Branching and interacting particle systems approximations of Feynman–Kac formulae with applications to nonlinear filtering. In Séminaire de Probabilités, XXXIV. Lecture Notes in Math. 1729 1–145. Springer, Berlin. MR1768060 10.1007/BFb0103798
[11] Douc, R., Fort, G., Moulines, E. and Priouret, P. (2009). Forgetting the initial distribution for hidden Markov models. Stochastic Process. Appl. 119 1235–1256. MR2508572 10.1016/j.spa.2008.05.007[11] Douc, R., Fort, G., Moulines, E. and Priouret, P. (2009). Forgetting the initial distribution for hidden Markov models. Stochastic Process. Appl. 119 1235–1256. MR2508572 10.1016/j.spa.2008.05.007
[12] Douc, R., Garivier, A., Moulines, E. and Olsson, J. (2011). Sequential Monte Carlo smoothing for general state space hidden Markov models. Ann. Appl. Probab. 21 2109–2145. MR2895411 10.1214/10-AAP735 euclid.aoap/1322057317
[12] Douc, R., Garivier, A., Moulines, E. and Olsson, J. (2011). Sequential Monte Carlo smoothing for general state space hidden Markov models. Ann. Appl. Probab. 21 2109–2145. MR2895411 10.1214/10-AAP735 euclid.aoap/1322057317
[13] Douc, R., Guillin, A. and Najim, J. (2005). Moderate deviations for particle filtering. Ann. Appl. Probab. 15 587–614. MR2114983 10.1214/105051604000000657 euclid.aoap/1107271661
[13] Douc, R., Guillin, A. and Najim, J. (2005). Moderate deviations for particle filtering. Ann. Appl. Probab. 15 587–614. MR2114983 10.1214/105051604000000657 euclid.aoap/1107271661
[14] Douc, R. and Moulines, E. (2008). Limit theorems for weighted samples with applications to sequential Monte Carlo methods. Ann. Statist. 36 2344–2376. MR2458190 10.1214/07-AOS514 euclid.aos/1223908095
[14] Douc, R. and Moulines, E. (2008). Limit theorems for weighted samples with applications to sequential Monte Carlo methods. Ann. Statist. 36 2344–2376. MR2458190 10.1214/07-AOS514 euclid.aos/1223908095
[15] Douc, R. and Moulines, E. (2012). Asymptotic properties of the maximum likelihood estimation in misspecified hidden Markov models. Ann. Statist. 40 2697–2732. MR3097617 10.1214/12-AOS1047 euclid.aos/1359987535
[15] Douc, R. and Moulines, E. (2012). Asymptotic properties of the maximum likelihood estimation in misspecified hidden Markov models. Ann. Statist. 40 2697–2732. MR3097617 10.1214/12-AOS1047 euclid.aos/1359987535
[18] Godsill, S. J., Doucet, A. and West, M. (2004). Monte Carlo smoothing for nonlinear time series. J. Amer. Statist. Assoc. 50 438–449. MR2054295 10.1198/016214504000000151[18] Godsill, S. J., Doucet, A. and West, M. (2004). Monte Carlo smoothing for nonlinear time series. J. Amer. Statist. Assoc. 50 438–449. MR2054295 10.1198/016214504000000151
[19] Gordon, N., Salmond, D. and Smith, A. F. (1993). Novel approach to nonlinear/non-Gaussian Bayesian state estimation. IEE Proc., F, Radar Signal Process. 140 107–113.[19] Gordon, N., Salmond, D. and Smith, A. F. (1993). Novel approach to nonlinear/non-Gaussian Bayesian state estimation. IEE Proc., F, Radar Signal Process. 140 107–113.
[20] Heine, K. and Crisan, D. (2008). Uniform approximations of discrete-time filters. Adv. in Appl. Probab. 40 979–1001. MR2488529 10.1239/aap/1231340161 euclid.aap/1231340161
[20] Heine, K. and Crisan, D. (2008). Uniform approximations of discrete-time filters. Adv. in Appl. Probab. 40 979–1001. MR2488529 10.1239/aap/1231340161 euclid.aap/1231340161
[22] Kingman, J. F. C. (1973). Subadditive ergodic theory. Ann. Probab. 1 883–909. MR356192 10.1214/aop/1176996798[22] Kingman, J. F. C. (1973). Subadditive ergodic theory. Ann. Probab. 1 883–909. MR356192 10.1214/aop/1176996798
[23] Kleptsyna, M. L. and Veretennikov, A. Yu. (2008). On discrete time ergodic filters with wrong initial data. Probab. Theory Related Fields 141 411–444. MR2391160 10.1007/s00440-007-0089-7[23] Kleptsyna, M. L. and Veretennikov, A. Yu. (2008). On discrete time ergodic filters with wrong initial data. Probab. Theory Related Fields 141 411–444. MR2391160 10.1007/s00440-007-0089-7
[24] Künsch, H. R. (2005). Recursive Monte Carlo filters: Algorithms and theoretical analysis. Ann. Statist. 33 1983–2021. MR2211077 10.1214/009053605000000426 euclid.aos/1132936554
[24] Künsch, H. R. (2005). Recursive Monte Carlo filters: Algorithms and theoretical analysis. Ann. Statist. 33 1983–2021. MR2211077 10.1214/009053605000000426 euclid.aos/1132936554
[25] LeGland, F. and Oudjane, N. (2003). A robustification approach to stability and to uniform particle approximation of nonlinear filters: The example of pseudo-mixing signals. Stochastic Process. Appl. 106 279–316. MR1989630 10.1016/S0304-4149(03)00041-3[25] LeGland, F. and Oudjane, N. (2003). A robustification approach to stability and to uniform particle approximation of nonlinear filters: The example of pseudo-mixing signals. Stochastic Process. Appl. 106 279–316. MR1989630 10.1016/S0304-4149(03)00041-3
[26] Le Gland, F. and Oudjane, N. (2004). Stability and uniform approximation of nonlinear filters using the Hilbert metric and application to particle filters. Ann. Appl. Probab. 14 144–187. MR2023019 10.1214/aoap/1075828050 euclid.aoap/1075828050
[26] Le Gland, F. and Oudjane, N. (2004). Stability and uniform approximation of nonlinear filters using the Hilbert metric and application to particle filters. Ann. Appl. Probab. 14 144–187. MR2023019 10.1214/aoap/1075828050 euclid.aoap/1075828050
[27] Olsson, J. and Rydén, T. (2008). Asymptotic properties of particle filter-based maximum likelihood estimators for state space models. Stochastic Process. Appl. 118 649–680. MR2394847 10.1016/j.spa.2007.05.007[27] Olsson, J. and Rydén, T. (2008). Asymptotic properties of particle filter-based maximum likelihood estimators for state space models. Stochastic Process. Appl. 118 649–680. MR2394847 10.1016/j.spa.2007.05.007
[28] Oudjane, N. and Rubenthaler, S. (2005). Stability and uniform particle approximation of nonlinear filters in case of nonergodic signals. Stoch. Anal. Appl. 23 421–448. MR2140972 10.1081/SAP-200056643[28] Oudjane, N. and Rubenthaler, S. (2005). Stability and uniform particle approximation of nonlinear filters in case of nonergodic signals. Stoch. Anal. Appl. 23 421–448. MR2140972 10.1081/SAP-200056643
[29] Pitt, M. K. and Shephard, N. (1999). Filtering via simulation: Auxiliary particle filters. J. Amer. Statist. Assoc. 94 590–599. MR1702328 10.1080/01621459.1999.10474153[29] Pitt, M. K. and Shephard, N. (1999). Filtering via simulation: Auxiliary particle filters. J. Amer. Statist. Assoc. 94 590–599. MR1702328 10.1080/01621459.1999.10474153
[33] Tadić, V. B. and Doucet, A. (2005). Exponential forgetting and geometric ergodicity for optimal filtering in general state–space models. Stochastic Process. Appl. 115 1408–1436. MR2152381 10.1016/j.spa.2005.03.005[33] Tadić, V. B. and Doucet, A. (2005). Exponential forgetting and geometric ergodicity for optimal filtering in general state–space models. Stochastic Process. Appl. 115 1408–1436. MR2152381 10.1016/j.spa.2005.03.005
[34] van Handel, R. (2009). The stability of conditional Markov processes and Markov chains in random environments. Ann. Probab. 37 1876–1925. MR2561436 10.1214/08-AOP448 euclid.aop/1253539859
[34] van Handel, R. (2009). The stability of conditional Markov processes and Markov chains in random environments. Ann. Probab. 37 1876–1925. MR2561436 10.1214/08-AOP448 euclid.aop/1253539859